www.digitalmars.com         C & C++   DMDScript  

digitalmars.D - D, Python, and Chapel

reply Russel Winder via Digitalmars-d <digitalmars-d puremagic.com> writes:
Chapel 1.11 just got release and they are making a big play on the=20
integration of Chapel with Python. This could be huge and potentially=20
disrupt the complacency of the NumPy based folk.

Chapel is a rather pleasant PGAS language that makes parallelism and=20
clustering quite nice. Certainly if the choice is Python+C++ vs=20
Python+Chapel, this is now a "no contest".

This may put a kibosh on the whole Python+D thing.

--=20
Russel.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder ekiga.n=
et
41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel winder.org.uk
London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder
Apr 03 2015
next sibling parent reply "Gary Willoughby" <dev nomad.so> writes:
Chapel overview: http://chapel.cray.com/overview.html
Apr 03 2015
parent David Gileadi <gileadis NSPMgmail.com> writes:
On 4/3/15 6:36 AM, Gary Willoughby wrote:

 Chapel overview: http://chapel.cray.com/overview.html


Their hello world examples do a fantastic job of illustrating their main 
selling point. My hat's off to whoever put that on their site.

D may have difficulty coming up with something like that, since its 
selling point(s) may be harder to distil. But if we could, it should get 
front 'n' center attention on dlang.org.
Apr 03 2015
prev sibling next sibling parent reply "John Colvin" <john.loughran.colvin gmail.com> writes:
On Friday, 3 April 2015 at 10:18:11 UTC, Russel Winder wrote:

 Chapel 1.11 just got release and they are making a big play on 
 the
 integration of Chapel with Python. This could be huge and 
 potentially
 disrupt the complacency of the NumPy based folk.

 Chapel is a rather pleasant PGAS language that makes 
 parallelism and
 clustering quite nice. Certainly if the choice is Python+C++ vs
 Python+Chapel, this is now a "no contest".

 This may put a kibosh on the whole Python+D thing.


I've had a look at Chapel and I don't get what the big deal is. 
There's some nice syntax and good thinking about parallelism in 
there*, but I don't see what's exciting after that... Maybe D has 
spoiled me for seeing power in a language.

I guess what I'm saying is I can see that they've put a lot of 
thought in to good abstractions for parallelism in HPC, we should 
steal a bunch of it because D is eminently capable of supporting 
similar abstractions, while being a much more rounded language in 
other regards.
Apr 03 2015
next sibling parent reply "Paulo Pinto" <pjmlp progtools.org> writes:
On Friday, 3 April 2015 at 15:34:05 UTC, John Colvin wrote:

 On Friday, 3 April 2015 at 10:18:11 UTC, Russel Winder wrote:

 Chapel 1.11 just got release and they are making a big play on 
 the
 integration of Chapel with Python. This could be huge and 
 potentially
 disrupt the complacency of the NumPy based folk.

 Chapel is a rather pleasant PGAS language that makes 
 parallelism and
 clustering quite nice. Certainly if the choice is Python+C++ vs
 Python+Chapel, this is now a "no contest".

 This may put a kibosh on the whole Python+D thing.


 I've had a look at Chapel and I don't get what the big deal is. 
 There's some nice syntax and good thinking about parallelism in 
 there*, but I don't see what's exciting after that... Maybe D 
 has spoiled me for seeing power in a language.

 I guess what I'm saying is I can see that they've put a lot of 
 thought in to good abstractions for parallelism in HPC, we 
 should steal a bunch of it because D is eminently capable of 
 supporting similar abstractions, while being a much more 
 rounded language in other regards.




The big deal is that is being developed in open collaboration 
with most companies and research labs that matter in HPC.
Apr 03 2015
parent Russel Winder via Digitalmars-d <digitalmars-d puremagic.com> writes:
On Fri, 2015-04-03 at 17:11 +0000, Paulo Pinto via Digitalmars-d wrote:

=20


[=E2=80=A6]

=20
=20
 The big deal is that is being developed in open collaboration=20
 with most companies and research labs that matter in HPC.


As is X10.

Cray and Washington University (Chapel) vs. IBM (X10)

It's sad that Fortress got canned, that was a really interesting=20
player in the PGAS language game.
=20
--=20
Russel.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder ekiga.n=
et
41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel winder.org.uk
London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder
Apr 04 2015
prev sibling parent Russel Winder via Digitalmars-d <digitalmars-d puremagic.com> writes:
On Fri, 2015-04-03 at 15:34 +0000, John Colvin via Digitalmars-d wrote:
[=E2=80=A6]

=20
 I've had a look at Chapel and I don't get what the big deal is.=20


It's PGAS programming, so you control the whole parallel computation=20
in a single program. Though for clustering you may have to suffer MPI=20
and hence an element of SPMD nastiness. Also there can be problems=20
distinguishing cores, processors, and computers =E2=80=93 the hardware has=
=20
three levels of parallelism, PGAS only two.=20


 There's some nice syntax and good thinking about parallelism in=20
 there*, but I don't see what's exciting after that... Maybe D has=20
 spoiled me for seeing power in a language.


I believe it would be most useful to unpack this to try and decide=20
what D has right, what Chapel has right, what D has wrong and what=20
Chapel has wrong.

My thinking here is that Chapel seems to have everything needed for=20
the practicing (not practising, though maybe that as well) HPC=20
programmer. D has lots but the core parallelism story is classic=20
single bus and so cannot handle the core/processor/computer split that=20
is easy with Chapel.


 I guess what I'm saying is I can see that they've put a lot of=20
 thought in to good abstractions for parallelism in HPC, we should=20
 steal a bunch of it because D is eminently capable of supporting=20
 similar abstractions, while being a much more rounded language in=20
 other regards.


But what does D have that people using Chapel should be demanding?=20
This is not to tell Brad et al. what they need to look at, but that as=20
well, more to say, why would these people consider D if the=20
parallelism message of D was stronger.

--=20
Russel.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D
Dr Russel Winder      t: +44 20 7585 2200   voip: sip:russel.winder ekiga.n=
et
41 Buckmaster Road    m: +44 7770 465 077   xmpp: russel winder.org.uk
London SW11 1EN, UK   w: www.russel.org.uk  skype: russel_winder
Apr 04 2015
prev sibling parent "bioinfornatics" <bioinfornatics fedoraproject.org> writes:
On Friday, 3 April 2015 at 10:18:11 UTC, Russel Winder wrote:

 Chapel 1.11 just got release and they are making a big play on 
 the
 integration of Chapel with Python. This could be huge and 
 potentially
 disrupt the complacency of the NumPy based folk.

 Chapel is a rather pleasant PGAS language that makes 
 parallelism and
 clustering quite nice. Certainly if the choice is Python+C++ vs
 Python+Chapel, this is now a "no contest".

 This may put a kibosh on the whole Python+D thing.


wow this look nice.

as example:
Parallel I/O: parallel file I/O, scientific file formats, 
parallel disks, extend domain maps to support

exactly what I asked at Andrei Alexandrescu -> 
http://forum.dlang.org/post/scgpiczjgzzvvtrclxer forum.dlang.org
Apr 06 2015